Fate of nitrous oxide (N2O) in Finnish boreal forest and peatland soils

(1)

FATE OF NITROUS OXIDE (N

₂

O) IN FINNISH BOREAL FOREST AND PEATLAND SOILS

Sankari Rajasakaren Master of Science Thesis MSc degree in Environmental Biology University of Eastern Finland (Kuopio), Faculty of Science and Forestry May 2017

(2)

2

Abstract

Recent studies show that nitrous oxide (N2O) has a missing sink in soils but the potential of this sink of N2O is poorly understood. Denitrification has been shown to be the major consumption process of N2O and could result as a sink potential for N2O if the process is highly efficient. However, the process of denitrification is influenced by several factors such as oxygen, pH, temperature, nutrient availability, soil moisture content and soil type. The main aim of this thesis was to understand the mechanisms that influence N2O concentrations in boreal forest soils and peatlands and to evaluate their potential to act as sinks for N₂O. The boreal forest study site selected for this study is located in Heinävesi and the peatland site in Salmisuo. Soil samples from the organic and mineral layer of the boreal forest and the upper 0-10cm and lower 10-20cm of the peatlands were collected to understand the differences in N₂O concentrations within different soil layers. Furthermore, the factors affecting N₂O concentrations were investigated, namely; the role of oxygen and substrate availability (more specifically, NH4+

, NO3-

). The effect of different carbon substrates namely succinate, butyrate, formate, propionate and lactate on N2O concentrations were also investigated to understand if they stimulated distinct microorganisms which influenced net N2O consumption/production potential.

In anoxic treatment both boreal peat (Salmisuo) and forest (Heinävesi) soil showed a net consumption of N2O. However, there were differences in net N2O consumption potential between forest soils and peat as well as between the different layers within each site. In oxic treatment forest soils showed a net production of N2O in organic soil but near negligible production of N2O in mineral soils. However in peat, there was a net consumption of N2O in both peat layers. The net consumption potential of N2O in peat was smaller in oxic conditions than in anoxic conditions. The addition of different carbon substrates to the mineral (Heinävesi) forest soil and the lower 10-20cm peat (Salmisuo) soil resulted in similar trends with exception of formate. Although there were similar trends observed in both soils, the consumption or production potential depending on the relevant carbon substrate was comparatively larger in the lower 10-20cm (Salmisuo) peat.

This study indicates that the effect of oxygen on denitrification plays a crucial role in regulating N₂O fluxes in boreal soils and their potential to act as sinks for N₂O. These results

(3)

3 also indicate that boreal soils are sensitive to aerobic/anaerobic conditions which can influence their potential to act as a sink for N₂O.

(4)

4

Acknowledgements

This master’s thesis represents to me a significant milestone in my life. It has been a period of intense learning, not only in the scientific aspects but also on a personal level. My time at University of Eastern Finland (UEF) at Kuopio has been nothing short of amazing. It has been the place that has provided me with unique opportunities since I came as an undergraduate exchange student from Germany in the autumn of 2011. Throughout my years at UEF, I have met many remarkable people who have been invaluable in my academic journey.

I would first like to express my sincere appreciation and wholehearted gratitude to my principle supervisor Jenie Gil Lugo. Her invaluable guidance and moral support helped me to complete this master thesis successfully. I am very much obliged to her for listening to all my concerns and even providing me with emotional support when I went through rough patches in completing this thesis. Ms Lugo always answered all my queries and made me feel comfortable in exploring my subject matter in the manner I chose to. She was also an important source of inspiration for me in completing this thesis.

I would also like to thank my second supervisor Professor Toini Holopainen for her invaluable guidance. I am extremely lucky and grateful that I was able to attain her in the reading of this manuscript. Professor Holopainen has been a tremendous beacon of light for me during this journey and without her support I would not have been able to complete a bachelor or master’s thesis. Albert Einstein once said, ‘It is the supreme art of the teacher to awaken joy in creative expression and knowledge’. This quote succinctly and aptly explains the high esteem I have for Professor Holopainen.

I would also like to thank my primary 6 school teacher, Mrs Chu-Tan Kim Yang. She made it fundamentally possible for me to enter the best schools in Singapore which enabled me to enter university. She encouraged her students to dream big and dream the impossible. I will forever be grateful to her; for I have always applied all the lessons she has taught me in life. I also owe much of my excellence to her.

Finally, I would like to thank my family and friends. My parents have provided me with limitless support and confidence in me which I cherish so dearly. They took the plunge with me in this journey since I arrived in Europe as a young, naive 19 year old. Nothing could have been possible without their love and encouragement. I am aware every single day how lucky I am to have parents like them who have constantly supported my dreams. Since I was a child,

(5)

5 my brother Balamuralli has been my role model. He is another reason why I was able to study abroad and gave me ample confidence to overcome any obstacle. I could not have asked for a better person to inspire me throughout my life. I would also like to thank my fiancé, Pierre- Emamnuel Panouillot. He has been extremely supportive and helpful throughout this entire journey and I am lucky to have met him in Kuopio. His unyielding friendship, love and support have enabled me to succeed. I cherish the numerous moments where he cared for me and helped me in preparing this thesis. He has always been there in all my victories and defeats and I could not have asked for a better partner in life. I would also like to thank all my friends and colleagues at NorskHydro in Norway who have been incredible. They are not simply work colleagues, superiors or friends but they are family. Their support has made this thesis possible. I owe my eternal gratitude for everything they have done for me.

Dedicated to my parents, Mr and Mrs Rajasakaren who are the most supportive and loving parents.

(6)

6

1. Introduction

1.1 Nitrous oxide

Nitrous oxide (N2O) is a potent greenhouse gas (GHG) contributing to climate change with an atmospheric lifetime of 114 years (Forster et al., 2007). With its ozone-depleting properties (Ravishankara et al., 2009) and a global warming potential 300 times that of carbon dioxide (CO₂), N₂O is attributed to be the third most important contributor to radiative forcing (Forster et al., 2007). As N₂O plays an important role in the atmosphere, identifying the sources and sinks of this greenhouse gas is critical for understanding its implications across ecosystems. Global N₂O emissions are estimated to be 17.71 Tg N yr^-1 (Schlesinger, 2013).

The main sources of N₂O to the atmosphere are agricultural (2.8 Tg N yr^-1) and tropical rain forest soils (4.0 Tg N yr^-1) contributing around 40% of the total emissions. Since the pre- industrial era, global atmospheric N₂O concentrations have increased by about 20%, which has been attributed largely to the increased use of nitrogen-based fertilizers (IPCC, 2013). It has also been predicted that with continuous increase of N2O in the atmosphere, the significance of N2O to greenhouse effect and ozone depletion will be more severe in the twenty-first century (Ravishankara et al., 2009).

The most important and main sink for N2O is the photochemical reactions (12.5 Tg N yr^-1), which take place in the stratosphere (Crutzen and Oppenheimer, 2008). However, there is growing evidence that soils can also behave as sinks for N2O (Chapuis-Lardy et al., 2007;

Schlesinger, 2013). Nevertheless, further understanding is necessary to elucidate the potential of soils as a sink of N2O. Soils prone to high N2O sink activity are mostly located in northern regions (Kroeze et al., 2007). Although Schlesinger (2013) has estimated global N2O uptake in soils to be not higher than 0.3 Tg N yr^-1 which corresponds to global sinks being no greater than 2% of estimated sources of atmospheric N2O, the exact potential of soils as N2O sinks have yet to be exactly determined. Hence, by understanding the mechanisms that influence N₂O uptake in soils we can ascertain the atmospheric importance of this N₂O sink.

In a review investigating 100 different studies on natural or recovering ecosystems, measured uptake potentials of N₂O in soils ranged from 1.0µg Nm^-2h^-1 to 207µg Nm^-2 h^-1 with a median value of 4µg Nm^-2 h^-1with the highest consumption of N₂O observed in wetland and peatland ecosystems (Schlesinger, 2013). The net fluxes of N₂O indicate whether soils are a net source or sink with N₂O production and consumption primarily related to the processes of nitrification and denitrification, respectively. These in turn depend on environmental factors

(9)

9 regulating consumption and production. Recent studies show that N₂O has a missing sink in soils but the potential of sinks of N₂O have been poorly delineated. It is pertinent to note that denitrification process can result as a sink potential for N₂O if the process is efficient.

For sink activity to occur, the ambient N₂O concentration in the atmosphere has to exceed that in the soil gas phase. The N2O exchange between soils and the atmosphere can be explained by the equilibrium concept (Conrad and Dentener, 1999). At N2O compensation point, the concentration of N2O in the soil gas phase results in no net exchange of N2O between ambient atmosphere and soil gas phase. Hence, it can be concluded that the compensation point determines the direction of the flux at a given ambient atmospheric N2O concentration.

Compensation concentrations of N2O are typically significantly higher than the ambient atmospheric concentrations which correlate with the observation that most soils are sources of atmospheric N2O. Studies that indicate consumption of N2O by soils suggest that during uptake the compensation concentrations are lower than ambient concentrations, i.e. <310ppbv (Kroeze et al., 2007). It has also been suggested that the compensation concentration for N2O increases with other factors such as temperature, soil moisture and nitrogen availability (Conrad and Dentener, 1999).

1.2 Research Objectives and Hypotheses

In this study, the main objective was to quantify the N₂O fluxes at a boreal spruce forest and pristine peatland in Finland in anaerobic and aerobic conditions. Two main hypotheses were tested. The first was that aerobic conditions in forest soils would result in markedly increased N2O fluxes in organic soils compared to mineral soils; but in both peat layers would result in consistent N2O fluxes. The second hypothesis was that the addition of different carbon sources to both forest soils and peat would affect N2O fluxes because of the resulting preference for certain carbon sources by microbes in the soils. To test these hypotheses experimental anaerobic and aerobic conditions and manipulation of carbon sources were arranged in flask microcosms and following N2O fluxes were determined. Furthermore, mineral nitrogen (nitrate and ammonium) concentrations in soil were quantified.

(10)

10 Figure 1: Circumpolar Range of Boreal Region (indicated in

green) (Hare and Ritchie, 1972)

2. Literature Review

2.1 Boreal Region

The boreal region is the biggest terrestrial biome in the world forming a distinct band of vegetation. It comprises nearly 14.5% of the Earth’s terrestrial ecosystems (Berner et al., 2015). The boreal region (Figure 1) is a circumpolar range throughout the entire Northern Hemisphere from Alaska and Canada; across the Atlantic to Norway, Sweden and Finland, through vast areas of Russia, and into parts of China, Mongolia, the Korean Peninsula and northern Japan (Miyawaki and Tüxen, 1977). It is characterized by harsh winters with mean temperatures below freezing for up to six months where much of boreal life comes to a halt (Runesson, 2011). Meanwhile, summers in boreal forests are characterized by long, warm periods of sunlight which leads to swift growth in boreal forests (Rumney, 1968 and Larsen, 1980). In Finland, the biogeography is composed of extensive boreal forests, mires and peatlands.

The boreal forests that dominate Finland are characterized by Pinus Silvestris (Pine), Picea abies (Spruce) and Betula pendula and B. Pubescens (Birch) which together encompass 73%

of Finnish land surface (METLA, 2013). The boreal forests are dominated by nutrient-poor

(11)

11 soils, with slow nutrient cycling. This is because of limited availability of nitrogen due to low mineralization rate and the high demand for available nitrogen for growth (Näsholm et al., 1998). Although the growth of boreal forests is generally N-limited, wide ranges in nutrient availability and interaction between elements have been observed (Vitousek et al., 1997).

Hence, it is important to understand nitrogen cycling processes thoroughly.

Peatlands are defined as a subset of wetlands containing peat-covered terrain having a minimum depth of peat of 30cm (Joosten and Clarke, 2002). Peat is a soil type, consisting of partially decomposed organic matter accumulating under water-saturated conditions due to incomplete decomposition. Peat forms as a result of anoxic condition, low decomposability of plant material and other factors. This accumulation of peat results in more acidic and nutrient- poor conditions because the influence of cations derived from mineral soil decreases over time (Vitt et al., 2005)

There are two main types of peatlands, minerotrophic and ombrotrophic. In minerotrophic peatlands, also known as fens, water in the peat surface is connected with, or has passed over or through surrounding mineral soil (Rydin and Jeglum, 2013). Fens have slightly acidic to slightly basic mineral-rich waters from groundwater discharge and seepage. They may be flat or gently sloping and are dominated by sedges, grasses, and non-sphagnum mosses. Fens can be classified in poor fen, intermediate-moderately rich fen and extremely rich fen based on their vegetation composition, local geology, chemical and nutrient status of the water supply among others. Poor fens have a pH of 4-5.5, while intermediate-moderately rich fens have a pH of 5-7, and rich fens have a pH of 6.8-8. On the other hand, ombrotrophic peatlands, also known as bogs, are isolated from mineral soil water and only receive their water and nutrients from precipitation. Bogs are nutrient poor and usually have a pH range of 3.5-4.2. They are dominated by the species of moss genus Sphagnum and heath shrubs (Rydin and Jeglum, 2013). Fens have a broader range of nutrient and pH variation and therefore more abiotic and biotic variation than bogs (Rydin and Jeglum, 2013).

In Finland, peatlands typically contain large amounts of carbon (C) and nitrogen (N) in organic matter in soil and vegetation (Gorham, 1991). Hence, they play an important role in the global C and N cycles particularly with processes related to producing and consuming greenhouse gases (GHG) (Khalil, 1999).

(12)

12

2.2 Processes that result in production and consumption of N

2

O in soils

N2O emissions from the soil can be largely attributed to bacterial and fungal respiratory processes in soils, namely denitrification and nitrification (Lassey and Harvey, 2007). About 70% of global N2O emission have been attributed to microbial processes of nitrification and denitrification in soils (Braker and Conrad, 2011; Syakila and Kroeze, 2011). However, it is impossible to solely ascribe the sources of N2O to microbial nitrification and denitrification because several biogeochemical pathways and abiotic mechanism produce or consume N2O (Butterbach-Bahl et al., 2013). Overall, soils are a complex heterogeneous matrix where N2O production can occur through different processes simultaneously in different microsites of the same soil (Butterbach-Bahl et al., 2013).

During the process of nitrification, there is microbial oxidation of ammonium (NH4+

) to nitrate (NO₃^-) by chemoautotrophic nitrifying bacteria in aerobic conditions. This process is affected by numerous factors including NH₄⁺, nitrite (NO₂^-), phosphate (PO₄^3-) concentrations, oxygen (O₂), pH, temperature, water potential and possible allelopathic compounds (Haynes, 1975). In aerobic autotrophic nitrification, there is stepwise oxidation of ammonia (NH₃) to hydroxylamine (NH₂OH) and subsequently to NO₂^-and NO₃^- (Kowalchuk and Stephen, 2001). N₂O production from these processes can result because of enzymatic decompositions of the substrates NH₂OH and NO₂^-(Arp and Stein, 2003; Baggs, 2011).

Biosynthesis of the intermediate NO₂^-as substrate for soilN₂O production can occur during the process of nitrate ammonification and nitrifier denitrification (Poth and Focht, 1985;

Baggs, 2011; Wrage et al., 2001). Chemodenitrification also contributes to N2O production when nitrite ion is chemically reduced to N2O by compounds such as amines present in soil organic matter, and by inorganic ions (Fe²⁺, Cu²⁺), particularly in subsoils (Granli and Bøckman, 1994).

Denitrification is a form of anaerobic respiration whereby the sequential dissimilatory reduction of NO3-

to NO2-

, nitric oxide (NO), N2O and dinitrogen (N2)is catalyzed by nitrate reductases (NaR or Nap), nitrite reductase (NiR), nitric oxide reductase (NoR) and nitrous oxide reductase (N2OR) respectively (Zumft, 1997). Denitrification is traditionally considered to require completely anoxic conditions because O2 is seen as a competing electron acceptor for NO3-

respiration and key enzymes of the denitrification pathways are inhibited by relatively small amounts of O2 (Zumft, 1997; Shapleigh, 2006). It is dominated by heterotrophic bacteria and fungi with the process considered to be the main N₂O producing process (Skiba and Ball, 2002). It is mainly affected by the presence of bacteria possessing

(13)

13 metabolic capacity; the availability of suitable reductants such as organic carbon; the restriction of O₂availability and the availability of N oxides such as NO₃^-, NO₂^-, NO or N₂O.

Closely coupled with denitrification, microbial co-denitrification can utilize N from organic co-substrates for increased N2O production (Kim et al., 2012).

In recent greenhouse gas models, the final step of denitrification (N2O  N2) has been assumed to be the major controlling mechanism reducing N2O fluxes to the atmosphere (Sanford et al., 2012). Most N2O is produced when there is incomplete denitrification in suboxic and anoxic conditions whereby anaerobic microbial bacteria reduce nitrate to nitrite and subsequently to gaseous nitrogen compounds nitric oxide (NO) and N2O. In table 1, adapted from Firestone and Davidson (1989), the factors that affect the proportion of N2O produced relative to N2 in denitrifying cells and soils are shown.

Table 1: Factors affecting the proportion of N2O and N2 produced during denitrification as adapted from Firestone and Davidson (1989)

Factor Will increase N₂O/N₂ [NO₃^-] or [NO₂^-] Increasing oxidant

O₂ Increasing O₂

Carbon Decreasing C availability

pH Decreasing pH

[H₂S] Increasing sulphide Temperature Decreasing temperature Enzyme status Low N₂O reductase activity

For N₂O to be produced, the availability of oxidant (N-oxide) has to exceed the availability of reductant (organic carbon) resulting in higher proportion of N₂O to N₂(Table 1). Moreover, decreasing temperature and lower N2O reductase activity also results in the same phenomenon (Table 1). The ratio of N2O/NO3-

has been observed to increase with lowering of oxygen partial pressure and increasing acidity (Goreau et al., 1980; Martikainen, 1985).

Although denitrification is traditionally believed to be inhibited by O2, there is substantial evidence for co-respiration or co-metabolism of O2 and NO3-

by different genera of cultured bacteria within the Alpha-, Beta- and Gammaproteobacteria under both alternating oxic/anoxic phases and fully aerated conditions (Robertson and Kuenen, 1984; Lloyd, et al.,

(14)

14 1987; Bonin and Gilewicz, 1991; Baumann et al., 1996). In bacterial cultures, N₂O and N₂ are produced when O₂ concentrations range from near anoxic to c.90% air saturation (Robertson and Kuenen, 1984; Körner and Zumft, 1989). N₂O produced during nitrification and denitrification contains some differences. N2O produced during nitrification is more depleted in ¹⁵N and ¹⁸O relative to substrates while N2O produced during denitrification is less depleted (Butterbach-Bahl et al., 2013).

Nitrogen oxides can be reduced by bacteria, archaea and fungi groups (Philippot et al., 2007;

Hayatsu et al., 2008). These micro-organisms can be classified as complete denitrifiers and non-denitrifying N2O reducers. Complete denitrifiers are facultative aerobes which are ecophysiologically homogenous with the potential to switch from oxidative respiration to denitrification when soil conditions are anoxic (Sanford et al., 2012). Non-denitrifying N2O reducers are denitrifiers with genes for N2O reduction only with atypical NosZ gene which encodes for N2O reductase (N2OR) and are more diverse and tolerate a wider range of conditions such as anoxic, microaerophilic, oxic, psychrophilic, piezophilic, thermophilic, and halophilic (Sanford et al., 2012). All denitrifiers in natural environments are assumed to be capable of complete denitrification resulting in N₂ as the final product (Firestone and Davidson, 1989). Bacteria and archaea containing the N₂OR also have the potential to convert N₂O to N₂ (Jones et al., 2011).

2.3 Atmospheric N

2

O fluxes from boreal soils

Microbial activity and abiotic environmental factors are important prerequisites for the efficiency of the N cycle in most terrestrial ecosystems, especially in N-limited ecosystems.

N2O fluxes from boreal forests are generally far less than those of tropical ecosystems and have perceived capabilities to have a much greater potential in these regions to act as a sink for N2O (Kroeze et al., 2007). In a study conducted in eastern Finland on afforested organic agricultural soils and organic agricultural soils in active use, the measured N2O flux values were in the range of 100-2500µg N2O-N m^-2 h^-1 with intermittent uptakes at sites with high water table level (Maljanen et al., 2012). This is probably due to the fact that the boreal forests are in regions of low temperatures and produce little N2O with emission rates less than 0.10 Tg N per year (Zhuang et al., 2012). However, in boreal soils from southern Russia and Canada N₂O emissions have been found to be relatively high at rates above 0.20 kg N₂O-N ha^-1 yr^-1 and this is primarily due to high soil organic matter content and moist climate (Zhuang et al., 2012).

(15)

15 Some studies have also shown N₂O uptake in soils in the boreal region. Net N₂O uptakes of up to -77µg N₂O-N m^-2 h^-1 has been measured from drained and abandoned, drained and afforested and active peat extractions in Finland. This uptake has been observed when soil pH and mean soil C:N ratio ranged between 3.9-5.9 and 17.6-24.2 respectively; with higher N2O uptake in soils with highly saturated water tables (Maljanen et al., 2012). This is because N- limited and high moisture ecosystems such as boreal mires and pristine boreal peatlands support N2O uptake (Chapuis-Lardy et al., 2007). It is however important to note that there are huge uncertainties involving boreal N2O fluxes due to high spatial and temporal variability (Ambus et al., 2006; Kellmann and Kavanaugh, 2008).

The highest spatial variability in soil N2O fluxes is always found in areas known as hot spots (Savage et al., 2014). “Hot spots” are areas that display disproportionately high reaction rates of biogeochemical activity relative to the surrounding area (McClain et al., 2003). Hot spots are often chemically limited because one reactant is unstable in the dominant biogeochemical environment, or because the reaction only proceeds under particular/specific conditions (for example, anoxia) (McClain et al., 2003). Biogeochemical activity can be further categorized as “hot” in the temporal dimension also known as “hot moments” whereby there are periods of intense biogeochemical reactions for a short period of time relative to longer intervening time periods. Hot moments can be delineated at time scales ranging from instants to millennia (McClain et al., 2003). For example, McClain et al., (2003) reported that denitrification hot spots within unsaturated soil profiles become active during hot moments in unsaturated zones where hydrological flow paths are intermittent with strong seasonal variations.

Denitrification hot spots have been identified across spatial scales range from soil profiles to larger basins. In a soil profile of the scale 1-10m, denitrification hot spots have been observed to occur around patches of labile organic matter, for example, plant detritus or manure (Christensen et al., 1990; Petersen et al., 1996); at the anaerobic center of large soil aggregates (Seech and Beauchamp 1988; Højberg et al., 1994); or in earthworm casts (Svensson et al., 1986; Parkin and Berry 1994). These hot spots occur because of the movement of nitrate-rich water into an organic-reducing substrate (Parkin, 1987; Johnston et al., 2001). Consequently denitrification hot spots require oxic-anoxic interfaces meeting a continual water flow, whereby oxic conditions are needed for nitrification to occur producing nitrate and anoxic conditions needed for denitrification and water acting as a transport medium (McClain et al., 2003).

(16)

16

2.4 Factors affecting N

2

O sinks

There are several physio-chemical factors affecting the potential for soils to serve as N2O sinks. These include soil oxygen, soil pH, soil temperature, nutrient availability, soil moisture content and soil type.

2.4.1 Effects of soil oxygen on N2O fluxes

Soil oxygen is important in controlling N turnover through the processes of nitrification and/or denitrification (Li et al., 2000 and Schurgers et al., 2006). Aerobic soils are generally sources of N2O but small uptake rates have sometimes been observed in dry soils (Duxbury and Moiser, 1993) and wet grass pastures (Ryden 1981, 1983). In anaerobic soils, a higher possibility for complete denitrification from N₂O to N₂ occurs. Therefore anaerobic soils are generally assumed to be the main potential sinks for N₂O (Erich et al., 1984). Despite this assumption, there has been no large, constant N₂O uptake observed in soils. For example, in flooded rice fields N₂O uptake has been largely dependent on the time of cropping seasons (Minani and Fukushi, 1984 and Parashar et al., 1991). The observed small N₂O uptake in flooded rice fields can be attributed to slow rates of dissolution and transport of atmospheric N2O in wet soils and hence prevent these sites from being a significant regulator of N2O emissions (Mosier et al., 1998).

In most surface soils, particularly in fertilized soils containing sufficient organic carbon and nitrate, the presence of oxygen most commonly limits denitrification (Firestone and Davidson, 1989). However, denitrification is extremely variable in space and time making it difficult to accurately ascertain rates of denitrification and N2O exchange in the field (Lapitan et al., 1999; Hofstra and Bouwman, 2005). N2O was found to be produced and consumed in both aerobic and anaerobic conditions when oxygen concentrations in soils were primarily affected by soil moisture content (Martikainen et al., 1993; Aerts and Ludwig, 1997). In oxygen- limited conditions, N2O becomes the sole electron acceptor for denitrifying microbes thus altering N₂O fluxes in soils (Butterbach-Bahl et al., 2013).

2.4.2 Effects of soil pH on N2O fluxes

Under field conditions, denitrification rates have been found to be lower in acidic soils compared to neutral or slightly alkaline soils (Kroeze et al., 2007). Simek and Cooper (2002) have described the amount of N2O relative to N2 formed during denitrification as pH dependent, mostly decreasing when pH increases. Hence, it can be deduced that the highest rates of N2O reduction occurs in neutral to alkaline soils. This could be explained by N2O reductase being hindered by a low pH (Richardson et al., 2009).

(17)

17 Moreover, soil pH is a major factor in the process of nitrification either directly or through its effects on soil cation exchange capacity (Robertson, 1989). Optimum pH values for nitrification are noted to range from 6.5 to 8 (Simek and Copper, 2002). When nitrate production increases through nitrification in soils at neutral pH, it could lead to a limited reduction of atmospheric N2O (Kroeze et al., 2007). According to Maljanen et al., (2012), sites with slightly higher mean soil pH (4.9) produced lower N2O than soils with slightly lower mean soil pH (4.4)

2.4.3 Effect of soil temperature on N2O fluxes in natural ecosystems

Soil temperature is a significant factor affecting denitrification which is one of the major processes affecting N2O fluxes. Typical temperature coefficient (Q10) values for denitrification range from 5 to 16 (Ryden, 1983) and the ratio of N₂O to N₂ produced decreases with increasing temperature (Firestone and Davidson, 1989). Sommerfeld et al., (1993) observed that N₂O emissions were high during the warm growing season in a temperate climate, and this was attributed to high microbial activity. According to Butterbach- Bahl (2013), many microbial processes in the nitrogen cycle are temperature-sensitive. Hence when temperature increases, soil respiration increases leading to decreased soil oxygen concentrations and increased soil anaerobiosis. Therefore, it might be possible that the potential for N₂O uptake through denitrification increases at higher temperatures. However, there have also been some contradictory results showing high N2O emissions at low soil temperatures during the winter and also during freezing and thawing events (Papen and Butterbach-Bahl, 1999; Maljanen et al., 2007).

2.4.4 Effect of nutrient availability on N2O fluxes

N-limitation in soils has been shown to cause intermittent N2O uptake (Rosenkranz et al., 2006; Vanitchung et al., 2011). In an acacia reforestation site with Acacia mangium trees, emission of N₂O was observed possibly due to the N-fixation activity of the trees which provided extra nitrogen to the soil (Vanitchung et al., 2011). Higher N2O emissions have also be observed in nitrogen fertilized cropping systems and also in tree-based systems where litter fall provide additional nutrient availability (Palm et al., 2002).

Atmospheric nitrogen deposition has been found to positively correlate with N₂O emissions (Butterbach-Bahl et al., 1998 and Zechmeister-Boltenstern et al., 2002). Even minor increases in deposition rates over extended durations may lead to changes in the nitrogen cycle of sensitive ecosystems and affects N2O sink potentials in soils (Bobbink et al., 1998;

Bouwman et al., 2002). This is largely because nitrogen enrichment of ecosystems partly

(18)

18 suppresses nitrogen limitation resulting in a reduced N₂O sink strength. According to Kroeze et al., (2007) even when sink activity is seasonal, this reduction of sinks of N₂O may be important considering the extensive areas of nitrogen-affected ecosystems which may in turn lead to increasing atmospheric N2O concentrations. Therefore, the total availability of nitrogen can also be a major driver of N2O soil emissions (Butterbach-Bahl et al., 2013).

High nitrate availability has been usually associated with high N2O emissions (Maljanen et al., 2012). In temperate regions, substrate accumulation in small water films in soils has also been known to result in high N2O emissions (Papen and Butterbach-Bahl, 1999; Teepe et al., 2000). Organic soils rich in C substrates have been found to be a significant source of N2O emissions in boreal regions (Alm et al., 2007). Blicher-Mathiesen and Hoffmann (1999) have noted that high nitrate concentrations commonly inhibit N2O reductase activity which elucidates the strong correlation between nitrate availability and N2O build-up observed in soils (Skiba et al., 1998). This supports the observation of net uptake of N2O in grassland and forest soils where nitrate concentrations are typically low (e.g. <1-2 µg NO3-

B/g soil;

Butterbach-Bahl et al., 1997). Nitrate concentrations are predicted to be lowest in environments with limited use of N-fertilizers and/or high plant uptake of nitrogen and also where nitrification does not occur (e.g. lack of oxygen and ammonia) (Kroeze et al., 2007).

Hence, this results in limited amount of electron acceptors in these environments and N₂ will be predominantly evolved. Moreover, in soils with an absence of oxygen, adequate organic carbon is present to support N₂O reduction (International Plant Nutrition Institute, 2016).

2.4.5 Effects of soil moisture content

One critical factor that affects boreal N2O fluxes is soil moisture content. This can be assessed differently e.g. as precipitation (Werner et al., 2007), water-filled pore space (WFPS) (Davidson et al., 2000) or water table level (WTL) (Maljanen et al., 2012). At above 60%

WFPS denitrification will occur because there is no absolute anaerobic situation and N2O can be produced as a by-product but when WFPS percentages are higher, the anaerobic situation is more pronounced and the production of N2O will decrease (Davidson et al., 2000).

Studies in boreal ecosystems showed that raising the water level close to the soil surface is likely to reduce N2O emissions with N2O emissions possibly mitigated when WTL is lower than 70cm (Maljanen et al., 2012). Thus, it can be deduced that soil moisture content is closely coupled with soil microbial activity which consequently affects nitrification and denitrification determining N₂O fluxes. High soil moisture content results in anoxic conditions which amplify N₂O production through denitrification (Vanitchung et al., 2011).

(19)

19 Soil moisture content also exerts considerable influence on activities of soil microbes, delivery of electron donors (NH₄⁺, DOC) and electron acceptors (O₂, NO₃^-) and in the diffusion of N trace gases from soils (Firestone and Davidson, 1989). It also regulates the availability of oxygen to soil microbes; and high water level in the soil enables high microbial N turnover rates ensuring that there are substrates available for soil microbes (Goldberg et al., 2010; Butterbach-Bahl et al., 2013).

In addition, WFPS might be linked with nitrate concentrations in affecting N2O emissions.

Limited nitrate conditions and higher WFPS stimulate denitrification enabling denitrifiers to utilize N2O as an electron acceptor and inhibiting nitrification (Vanitchung et al., 2011).

Moreover, increased N substrates and easily degradable C availability have also been found to increase microbial N2O emissions (Papen and Butterbach-Bahl, 1999).

2.4.6 Effect of soil type on N2O fluxes

2.4.6.1 Soil Carbon-to-Nitrogen ratio (C:N ratio)

Soil carbon-to-nitrogen ratio (C:N ratio) determines the decomposability of soil organic matter which in turn has a critical impact on soil nitrogen availability. Relationships between soil C:N might play a key role in nitrous oxide emissions. Soil residues with lower C:N ratios have been observed to have a higher decomposition rate hence providing more dissolved organic carbon (DOC) and therefore increasing N2O emissions (Huang et al., 2004). Soil C:N ratio affects the mineralization of plant residues and consequently N2O emissions (Aulakh et al., 1991 and Németh et al., 1996). Organic amendments to a well-aerated soil which decrease C:N ratios have been shown to increase N2O emissions (Bremmer and Blackmer, 1981).

Substrates with C:N ratios <20 have a higher decomposition rate resulting in ammonium release via mineralization increasing N2O emissions. Substrates with ratios of 25-75 also decompose quickly but N mineralization is inhibited by increased microbial immobilization and protein complexation by polyphenols when the cells lyse. In a modelling study in Scandinavia and the Baltic States, soils with elevated C stocks had considerably higher N₂O emissions than ambient soils; with values of >0.75 kg N ha^-1 being calculated (Kesik et al., 2005). High N₂O emissions were also predicted via modelling for soils with high amounts of organic carbon content in the forest floor in Southwest Finland and in the Northern parts of Sweden (1.0 to 1.8 kg N ha⁻¹ yr⁻¹) (Kesik et al., 2005).

(20)

20 2.4.6.2 N2O emissions from Organic vs mineral soil types

N2O emissions from mineral soils in boreal forests are in the range of 0.1 and 0.3 kg N ha-1 yr^-1 (Brumme et al., 2005), while forest soils, rich in organic matter, emitted N2O in the range of 1.0 to 10.0 kg N ha-1 yr^-1 (Maljanen et al., 2001, 2003; von Arnold et al., 2005). N2O emissions from peat soils (organic soil) that have been used for agriculture prior to forestation emitted the most N2O (Maljanen et al., 2003). Therefore, it can be concluded that large N2O emissions occur as a result of drainage and cultivation of organic soils due to enhanced mineralization of old, N-rich organic matter (Guthrie and Duxbury, 1978; Martikainen et al., 1996; Velthof et al., 1996). Nutrient poor organic forests were noted to emit negligible amounts of N2O (Regina et al., 1996). However, drained organic soils with no fertilizer additions were noted to show much higher emissions of N2O of up to 100 kg N- N2O ha-1 yr^-1 than mineral soils (Regina et al., 1996).

2.4.6.3 Natural vs altered ecosystems

Human activities like agriculture affect the nitrogen cycle by increasing N2O emissions. The production of synthetic nitrogen fertilizers and cultivation of N2-fixing plants play a key role in steadily increasing nitrogen into the biosphere thus altering the nitrogen cycle (Vitousek et al., 1997). Nitrogen fertilizers that contain ammonium and nitrate have been known to elevate the emissions of N2O immediately after addition (Eichner, 1990, Chang et al., 1998).

Drainage along with other agricultural practices such as ploughing, liming and fertilization contribute to higher pH values in soil and stimulate decomposition of N-rich organic matter (Maljanen et al., 2012) and nitrogen mineralization (Freeman et al., 1996).

It is known that emissions of N2O from pristine peatlands are negligible but in drained peatlands there is an increase in N mineralization leading to greater emissions of N2O due to nitrification and denitrification (Regina et al., 2004). In Nordic countries, agricultural activities conducted in organic soils resulted in N₂O emissions on average four times higher than those from mineral soils, indicating that N2O derived from soil organic carbon decomposition dominates overall fluxes (Mu et al., 2014). Moreover, in organic soils the variability of soil C/N ratio may be one of the dominant factors determining N₂O emissions in organic soils (Mu et al., 2014).

Several studies have noted that clayey and highly fertile soils result in higher N₂O emission (Matson and Vitousek, 1987; Verchot et al., 1999). Forest soils are more inclined to provide acidic conditions which may select microorganisms to different processes according to their tolerance of pH ranges (Muñoz et al., 2010). Acidic soil conditions of coniferous forests in

(21)

21 Western Europe contain heterotrophic bacteria and fungi promoting nitrification with the bacteria Arthrobacter sp. seemingly most highly adapted to initiate heterotrophic nitrification (Brierley and Wood, 2001).

In the forests of South-Central Chile, dominated by volcanic soils the N-cycle is extremely efficient resulting in low productions of N gases to the atmosphere due mainly to the physicochemical characteristics of these soils (Chorover, 2002; Godoy et al., 2003). In pristine soils, N conservation can be attributed to consumption by microorganisms and vascular plants by net primary production (Muñoz et al., 2010). There is a scarce production of nitrate in these soils resulting in nitrification requiring an extra-consumption of energy at the ecosystem level (Huygens et al., 2008). Hence, this results in an increased amounts of N immobilized by microbial action and adsorption of inorganic forms of N onto clay colloid surfaces (Bengtsson and Bergwall, 2000; Huygens et al., 2008). It has also been noted that land use change from native forest to forest plantations and grassland remarkably increased N mineralization and nitrification in soils of New Zealand (Parfitt et al., 2003).

2.4.7 Hole in the pipe model

Firestone and Davidson (1989) have proposed a model known as “hole-in-the-pipe” (HIP) which explains how microbiological and ecological factors affected soil emission of NO and N₂O fluxes. Several studies have indicated that nitrogen fertilization encouraged production of one or both gases (Williams et al., 1992). On the contrary, in unfertilized soils net nitrogen mineralization and net nitrification have been found to positively correlate with N2O emissions (Robertson and Tiedje, 1984, Matson and Vitousek, 1987). In the HIP model, the total production of NO and N2O is assumed to be directly linked to the availability of nitrogen in the soil. The HIP model also attempts to link soil water content to the ratios of N2O:NO emissions as a function of soil water content (Davidson et al., 2000).

The HIP model is expressed by leaky pipes representing two major processes nitrification and denitrification. The rate of flow of nitrogen through these pipes is comparable to the rates of nitrification and denitrification and shows nitrogen cycling through the ecosystem (Davidson et al., 2000). “Holes” in the pipe represent trace gases of NO and N₂O leaking out of the pipe and the sizes of these holes correspond primarily to soil water content (Davidson et al., 2000).

(22)

22

3. Materials and Methods

3.1 Study site

The first study site was located in eastern Finland in Heinävesi (62º26’N, 28º 38’E) which is an upland forest of Myrtillus type (MT) (Hotanen et al., 2008). The main tree species located at the site are spruce (Picea abies), pine (Pinus sylvestris) and birch (Betula pendula and Betula pubescens) with a stand age of 80 years and the dominant soil type in this area can be characterized by a humus layer and mineral soil.

The second study site was located in eastern Finland in Salmisuo, Ilomantsi (62°47′N, 30°56′E) in a low-sedge Sphagnum papillosum pine fen. The Salmisuo peatland complex is an eccentric bog with some minerotrophic strips. Plant species indicate average water tables and nutritional statuses of the mire types and microsites. Flarks are the wettest microsites, with an average water table depth from 0 to –5 cm while lawns have an average water level depth 5–

20 cm below the peat surface. Hummocks rise above their surroundings and have an average water table over 20 cm below the peat surface. The vegetation in Salmisuo enabled all three distinctive forms - hummocks, lawns and flarks, to be recognized in the low-sedge S.

papillosum pine fen.

The proportion of different microsites were determined by evaluating their coverage in 20 vegetation squares (1×1 m, along a transect with 5-m intervals). The most typical hummock species was Sphagnum fuscum (Schimp.) Klinggr., with S. angustifolium (Russow) C. Jens.

and Eriophorum vaginatum L. also found in the lower parts of the hummocks. The lawn moss layer was dominated by S. angustifolium and S. balticum (Russow) C. Jens., accompanied by some S. magellanicum Brid. and S. papillosum Lindb. The field layer in the lawns consisted mainly of E. vaginatum with Andromeda polifolia L., Vaccinium oxycoccos L. and Carex pauciflora Light. The major mosses in the flarks were S. majus (Russow) C. Jens., S. balticum and S. angustifolium, and the only vascular plants on flarks were Scheuchzeria palustris L.

and E. vaginatum. Between the mineral soil and the low-sedge S. papillosum pine fen there was a narrow tall-sedge fen lag consisting of a lawn vegetation in which the major moss species was S. angustifolium and the field layer was dominated by C. rostrata Stokes. In this study, upper peat layers of 0-10cm and lower peat layers of 10-20cm were sampled.

Organic and mineral soil at the Heinävesi site (Fig. 2) and profiles of 0-10cm depth and 10- 20cm depth at the Salmisuo site were collected on the 2^nd of June 2014 and the 8^th of July

(23)

23 Figure 2: Organic and mineral layers of forest soil at

Heinävesi site

2014 respectively. Soils were collected and sealed in ziplock bags and refrigerated in cold room of temperatures up to 4°C until physical and chemical analysis of soils were conducted.

Soils were then homogenized manually by picking out roots and debris. Soil pH was measured from soil-water suspension (1:2-3 v/v) by using a pH meter (WTW pH-Electrode SenTix® 81, Germany) while electrical conductivity was measured using an EC meter (WTW TetraCon® 325, Germany). Soil organic matter was also analyzed for the respective soil layers utilizing the weight loss on ignition method (Reeuwijik, 2002). Samples were first oven-dried at 105°C in a muffle furnace and thereafter the samples were burnt at 450°C (Reeuwijik, 2002). The weight loss that occurred at this temperature was then correlated to the soil moisture content.

3.2 Experimental set-up

3.2.1 Effect of oxygen on N2O, NH4+ and NO3- concentrations

The experiment was carried out in flask-microcosms (Figure 3) in oxic and anoxic conditions.

Sacrificial sampling was carried out during the experiment. To ensure that there was no N2O consuming activity already present, pre-incubation was done for 5 days whereby bottles were degassed and flushed with 100% helium. This degassing and flushing was done twice to ensure that there was complete removal of accumulated N₂O during the 5 days. After the pre- incubation period, N₂O consumption was activated in the respective soil layers (mineral/organic; upper 0-10cm/lower 10-20cm) of the Heinävesi and Salmisuo soils with

(24)

24 Figure 3: Flask-microcosm experiment (Peat samples)

addition of 15-N N₂O (98% atm, 1500 ppb) to the headspace of the bottles. The respective soil layers of each soil site were then subjected to oxic and anoxic conditions. Oxic conditions were achieved by adding 21% volume of O₂ to the headspace of the bottles while anoxic conditions did not have any additions of O2. Sampling of nitrous oxide gas, nitrate and ammonium was then done at various time points for each site (0, 3, 12, 48, 72, 168 hours for Heinävesi and 0, 3, 24, 48, 72, 168 hours for Salmisuo). There were five replicates for each treatment (anoxic/oxic conditions) and incubation experiments were conducted at room temperatures of 22°C.

N₂O concentrations were then determined by Gas Chromatography analysis (Hewlett Packard 5890 Series II, U.S.A). Ammonium-N (NH₄⁺-N) and nitrate-N (NO₃^-N) were extracted from integrated soil samples with 0.5M K₂SO₄ (soil:K₂SO₄ 1:2 v/v, 175 rev min-1, 1 hour) at various time points during the incubation period (168h). The extracts were filtered (Blauband 589³ BlueRibbon filter paper) overnight and stored at -20°C until analyzed for NH4+

and NO3-

. The concentration of NH4+

and NO3-

in the extracts were measured by UV-Visible spectrophotometry (Philips PU 87501 UV/VIS) using microtiter plate format and following the protocol of Fawcett and Scott (1960) for NH4+

(630 nm) and Griess method for NO3-

in aqueous solution (544 nm) (Miranda et al.,2001).

(25)

25 3.2.2 Effect of electron donors on mineral forest soil and lower 10-20cm peat soil on N2O concentrations in oxic conditions

We also wanted to understand the effects of electron donors on the N2O fluxes in mineral layer of Heinävesi and 10-20cm layer of Salmisuo sites. Five electron donors (succinate, butyrate, formate, propionate, lactate) were added to soils (Heinävesi: mineral; Salmisuo: 10- 20cm depth). There were 3 replicates of each treatment and 3 controls for each site. Soils were added to bottles and pre-incubated in anoxic conditions for five days at 14°C and degassed and flushed with 100% Helium. N₂O consumption was activated by adding 15-N N₂O (98% atm, 1500 ppb) to the headspace of the bottles. Oxic conditions were then ensured in the bottles by adding 21% volume of O₂to the headspace of bottles. This was then followed by the addition of 1mM of each electron donor to the respective bottles. Sampling of nitrous oxide was done at various time points (0, 3, 5, 24, 48, 72, and 168 hours) for both soils.

3.3 Statistical Analysis

Statistical analysis was performed with the IBM SPSS Statistics 21. The Shapiro-Wilk test of normality of variables was first carried out on all data to check for normality of data. As the data did not conform to normality, non-parametric tests (Kruskal-Wallis and Mann-Whitney U) were used to compare the data sets between sites over time.

(26)

26

4. Results

4.1 Soil properties of the study sites

Heinävesi forest soils have a considerably lower NH4+

content than Salmisuo peatland soils with the lowest NH4+

content observed in the mineral layer of Heinävesi soils and the highest NH₄⁺ content in the 10-20cm layer of the Salmisuo soil (Table 2). The same trend was observed for the NO₃^- concentrations, with the Heinävesi soils having a notably lower NO₃^- content than the Salmisuo soils. The lowest NO₃^- content was observed for the mineral layer of the Heinävesi soil but the highest NO₃^- content was observed in the upper (0-10cm) layer of the Salmisuo soil. For the Salmisuo soil the upper layer (0-10cm) had nearly twice as much NO₃^- than the lower layer (10-20 cm) (Table 2).

Table 2: Background Soil properties of the study sites

Site Soil

Type µg N-NH4/g_dw µg N-NO3/g_dw pH Temperature EC (µ S/Cm)

Soil Moisture (%) Heinävesi Mineral 2.78 ± 0.90 0.023 ± 0.003 4.89 21.1 15 8.31 Heinävesi Organic 5.30 ± 0.57 0.033 ± 0.008 4.68 21.1 14 31.71

Salmisuo 0-10 26.78 ± 0.26 0.87 ± 0.53 4.25 22.3 20 88.59

Salmisuo 10-20 33.62 ± 1.61 0.46 ± 0.13 4.76 22.3 18 90.76

Both soils had very similar pH range and were largely acidic. The upper layer (0-10cm) of the Salmisuo soil was the most acidic, followed by the organic layer of the Heinävesi soil, the lower layer of Salmisuo soil (10-20cm) and the mineral layer of Heinävesi respectively (Table 2). In general, the Salmisuo peat soil had higher soil moisture content than the Heinävesi upland forest soils. Both peat layers (Salmisuo soil) have similar soil moisture contents but the organic upland forest soil (Heinävesi soil) had higher soil moisture content than the mineral layer (Table 2).

(27)

27

4.2 Effects of Anoxic conditions on N

2

O fluxes in forest soils (Heinävesi)

When forest soils were exposed to anoxic conditions, N2O concentrations decreased in both mineral and organic soil layers (Figure 4a). A pronounce decrease in N2O concentrations in both soil layers was observed between 3 and 48 hours of incubation with the organic soil showing greater mean net consumption of N2O (decreased in concentration) than the mineral soil throughout the experiment (Figure 4a, Table 3). The effect of anoxia on this decrease in N2O was statistically significant during this time frame (Table 4). Anoxic conditions also significantly resulted in the near total consumption of the N2O in the organic layer by the end of the experiment (Figure 4a). The forest mineral layer however, was only able to reduce half of the N2O concentrations by the end of the experiment compared to initial N2O fluxes (Figure 4a).

In addition, a statistically significant increase in ammonium concentrations in the organic layer of the forest soil was measured throughout the experiment under anoxic conditions (Figure 4b, Table 5). For the mineral layer, the effect of anoxia was not significant since the concentrations of ammonium remained constant and negligible throughout the experiment (Figure 4b, Table 5). On the other hand, nitrate concentrations in both soil layers (organic and mineral) showed a slightly decreasing trend throughout the experiment (Figure 4c), but it was not statistically significant (Table 6).

(28)

28

0 1000 2000 3000 4000 5000 6000

0 3 12 48 72 168

µg N-N2O/gdw

Time (hrs)

Heinävesi Anoxic

Mineral Organic

0 5 10 15 20 25 30

0 3 12 48 72 168

µg N-NH4/gdw

Time (hr)

Heinävesi Anoxic

^Mineral

Organic

Figure 4a: Effect of anoxic conditions on N2O fluxes in Heinävesi soils

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50

0 3 12 48 72 168

ug N-NO3/gdw

Time (hrs)

Heinävesi Anoxic

_mineral

organic

Figure 4c: Effect of anoxic conditions on NO3-

concentrations in Heinävesi soils

Figure 4b: Effect of anoxic conditions on NH₄⁺ concentrations in Heinävesi soils

(29)

29 Table 3: Net N₂O consumption/production potential in Heinävesi and Salmisuo soils

in anoxic/oxic treatment

Site Soil Type Anoxic Treatment Oxic Treatment

Heinävesi Mineral -91.28 1.10

Heinävesi Organic -322.14 493.53

Salmisuo 0-10cm -187.38 -40.61

Salmisuo 10-20cm -319.06 -63.15

(30)

30 Table 4: Effect of oxygen on nitrous oxide (N₂O-N) concentrations (* indicating p<0.05)

Site Soil Time Significance

Heinavesi Mineral 0 0.221

3 0.016 *

12 0.009 *

48 0.009 *

72 0.047 *

168 0.117

Organic 0 0.329

3 0.009 *

12 0.016 *

48 0.009 *

72 0.009 *

168 0.009 *

Salmisuo

0-10cm 0 0.117

3 0.016 *

24 0.009 *

48 0.009 *

72 0.009 *

168 0.009 *

10-20cm 0 0.047 *

3 0.009 *

24 0.009 *

48 0.009 *

72 0.009 *

168 0.009 *

(31)

31 Table 5: Effect of oxygen on ammonium (NH₄⁺-N) concentrations (* indicating p<0.05)

3 0.465

12 0.068

48 0.602

72 0.475

168 0.076

Organic 0 0.034 *

3 0.028 *

12 0.009 *

48 0.009 *

72 0.009 *

168 0.009 *

Salmisuo

0-10cm 0 0.465

3 0.602

24 0.221

48 0.602

72 0.028

168 0.034

10-20cm 0 0.117

3 0.699

24 0.602

48 0.624

72 0.289

168 0.053

(32)

32 Table 6: Effect of oxygen on nitrate (NO₃^--N) concentration (* indicating p<0.05)

3 0.465

12 0.142

48 0.064

72 0.297

168 0.456

Organic 0 0.329

3 0.564

12 0.076

48 0.717

72 0.784

168 0.175

Salmisuo

0-10cm 0 0.251

3 0.465

24 0.245

48 0.472

72 0.327

168 0.251

10-20cm 0 0.754

3 0.347

24 0.602

48 0.251

72 0.347

168 0.754

Fate of nitrous oxide (N2O) in Finnish boreal forest and peatland soils